22 Fishery Board for Scotland. 



It is possible that one may be able to compare certain shoals by 

 the observed averages, or by combinations of averages. The curve 

 of probability is mainly of value for the calculation of the average, 

 when only a few fishes are available for examination. 



The means and fluctuations of the means should, however, also be 

 calculated. 



The Observer and the Personal Error. 



It would be well if each investigator limited his work to the herrings 

 frequenting one spawning-ground. He should make all the observa- 

 tions and measurements. He might receive assistance in the sub- 

 sequent mathematical calculations. The investigators should meet 

 and discuss the results. The relation between their personal errors 

 should then be determined by experiment. 



Literature. 



Heincke, F. — " Naturgeschichte des Herings. Teil I. Die 

 Lokalformen und die Wanderungen des Herings in den Europaischen 

 Meeren. Text, Tabellen, und Tafeln." Abhandlungen des Deutschen 

 Seeftscherei-Vereins. Bd. ii. Heft 1. Berlin. 1898. 



Jenkins, J. T. — " The Difference between Spring and Autumn 

 Herring." No. XI. Report for 1902 on the Lancashire Sea-Fisheries 

 Laboratory, etc. Liverpool. 1903. Pp. 28-38. 



Kyle, H. M.— " The Natural History of the Herring," by Fr. 

 Heincke. Abstract. 17th Annual Report of the Fishery Board for 

 Scotland for 1898. Pt. III. 1899. Pp. 274 and following. 



Matthews, J. D. — " Report as to Variety among the Herrings 

 of the Scottish Coasts." Part I. 4th Annual Report of the Fishery 

 Board for Scotland for 1885. (1886.) P. 61. Pt. II. 5th Annual 

 Report of the Fishery Board for Scotland for 1886. (1887.) P. 295. 



Explanation of Figures. 



Fig. 1. Diagram of herring, showing some external measurements. 



(After Heincke.) 

 ,, 2. Dorsal view of skull, to exhibit measurements. (After 



Heincke.) 

 ,, 3. Hind view of skull, to exhibit measurement. (After 



Heincke.) 



